33-41, 1983. RECENT STUDIES, both experimental'-5 and clinical,69 report the feasibility of recording an extracellular sinus node electrogram with a conventional transvenous electrode catheter technique.Experimental proof for the validity of sinus node electrograms was provided by simultaneously recording the transmembrane action potential and the extracellular potential of the sinus node and by inducing sinus nodal exit block with tetrodotoxin (TTX).2 The clinical evidence is more indirect and relies on the temporal relationship of the sinus node electrogram to the P wave and its similarity to the waveform recorded experimentally.We reasoned that if the periods of atrial quiescence were sufficiently long, we could record multiple, repetitive, isolated sinus node electrograms. This recording, which is possible in the long atrial pauses observed after pacing in patients with sick sinus syn-

No abstract

No abstract

fig. 2), accounting for a more than compensatory pause. The morphology of this SRE is exactly the same as during postpacing atrial pause as shown in the inset (note the different paper speeds); the morphology is different from that in figure 1 because the position of the sinus lead was slightly modified by this time.Likewise, during prolonged spontaneous alternation of atrial cycles, an identical SRE emerges before A at the end of each long atrial cycle, suggesting that alternating conduction times is the mechanism for the short-long cycle altermation. From these observations and again assuming conduction, we suggest the working hypothesis that the unique ability of extrasystoles to markedly prolong sinoatrial conduction time (SACT) up to the point of complete block characterizes human SSS. It would account for the more than compensatory pauses during programmed atrial stimulation in some patients and for the long postpacing atrial pause (abnormal CSNRT) by a cumulative effect in most patients. These findings do not contradict the observations of Gomes et al. that indirect SACT correlates with its direct measure because the CSNRT (which measures the maximuni depressive effect in our hypothesis) of their patients was much shorter than in ours (by design, because a longCSNRT is necessary to demonstrate multiple SREs). The mechanism of this postpacing block is unknown to us, but might be investigated in a dog model we have described.